Nondestructive Defect Detection in Castings by Using Spatial Attention Bilinear Convolutional Neural Network

X-ray images of castings are widely used in manufacturing for quality assurance. This article investigates the X-ray-image-based defective detection. The main contributions this twofold: first, a new full-image method is proposed to classify and nondefective ones; second, by combining two technologies, spatial attention mechanism bilinear pooling deep convolutional neural networks (CNNs), CNN enhance representation power CNN. To validate above initiatives, extensive experimental studies have been carried out show advantages over number existing ones.